{"action":"Upgrade llama.cpp to 7593+dfsg-1 or later.","affected":[{"fixed":"7593+dfsg-1","introduced":"0"}],"component":"llama.cpp","cve_ids":[],"ecosystem":"Debian","ecosystem_qualifier":"Debian:14","exploited":false,"fixed_versions":["7593+dfsg-1"],"id":"tcs_01M5NQZ00QQ05TBBA5TF23HKM0","modified_at":"2026-10-01T03:03:20.483150331Z","published_at":"2026-01-08T00:16:00.297Z","references":["https://osv.dev/vulnerability/DEBIAN-CVE-2026-21869","https://security-tracker.debian.org/tracker/CVE-2026-21869"],"severity":{"label":"critical","score":9.8},"sources":[{"source":"osv","source_id":"DEBIAN-CVE-2026-21869","url":"https://osv.dev/vulnerability/DEBIAN-CVE-2026-21869"}],"summary":"llama.cpp is an inference of several LLM models in C/C++. In commits 55d4206c8 and prior, the n_discard parameter is parsed directly from JSON input in the llama.cpp server's completion endpoints with"}